Our thinking

When AI goes unchecked

30 May, 2025

With summer in full swing, you might be looking for inspiration on what to eat, wear, or read. But before you dive into that trending summer reading list, double-check it’s real, as some AI-generated versions have included made-up books by famous authors. Planning to visit a buzzy new restaurant in London? Make sure it actually exists, even if it is trending on Google. And if your horoscope tells you to “avoid all triangles today,” maybe take that with a pinch of salt. 

These are the more amusing consequences of AI gone unchecked. But they sit on a spectrum that includes far more serious issues, like AI-generated medical misinformation, political deepfakes, and market manipulation. 

The need for stronger AI governance is becoming urgent. A new report from Sustainalytics shows that shareholders are stepping in, pushing Big Tech companies to adopt clearer oversight. Proposals calling for transparency, ethical guardrails, and board-level accountability are gaining traction, even as broader ESG enthusiasm slows. 

For companies, the risks are real: reputational damage, public mistrust, and operational disruption. For investors, these incidents are red flags. As AI becomes embedded in everything from customer service to content moderation, vague assurances no longer cut it. Shareholders want to know who’s responsible, how decisions are made, and what happens when things go wrong. 

This isn’t just about avoiding risk, it’s about building long-term value. Companies that treat AI oversight as a strategic priority, not a compliance exercise, are more likely to earn trust, attract talent, and stay ahead of regulation. 

We don’t have all the answers yet. But with growing shareholder momentum, boards are starting to ask the right questions. 

By Alice Railton